ClimateLearn: Benchmarking Machine Learning for Weather and Climate New

ClimateLearn (Nguyen et al., 2023; UCLA Machine Intelligence Group, led by Prof. Aditya Grover) is a Python library for accessing state-of-the-art climate data and machine learning models in a standardized, straightforward way. It offers multiple datasets, a zoo of baseline approaches, and a suite of metrics and visualizations for large-scale benchmarking of statistical downscaling and temporal forecasting methods.

The library lowers the barrier for researchers to reproduce and compare climate-ML results by unifying data loading, model definitions, and evaluation. Licensing: the code is released under the MIT License.

Why it matters for physics AI: alongside weather-focused models such as Pangu-Weather, FourCastNet, GenCast and NeuralGCM, ClimateLearn provides the evaluation scaffolding - datasets, baselines, and metrics - that makes climate and weather ML results comparable and reproducible, which is essential infrastructure for the field.